Erlotinib is a tyrosine kinase inhibitor (TKI) that binds to the epidermal growth factor receptor (EGFR). Erlotinib is an extremely effective therapeutic agent and is commonly prescribed as a first-line therapy for non-small cell lung cancer (NSCLC) provided that patients have an activating mutation of EGFR. However, erlotinib is not effective in all NSCLC with activating mutations. We have shown previously with PET imaging using the tracer 11C-erlotinib that only mutations of the kinase domain of EGFR convey sensitivity to erlotinib. Unfortunately even if initially effective, treatment with EGFR TKIs eventually gives way to acquired drug resistance. At present, there is no way to predict when resistance will occur. When it does, we do not know if the mechanism of resistance is equivalent at all tumor sites. With PET, we have the power to image the degree of erlotinib binding to its target site in tumors with mutant EGFR. Knowing the erlotinib-binding status of patients could be useful to physicians for predicting treatment success and for making treatment decisions. Our long term goals are to conduct a clinical trial to determine (a) if 11C-erlotinib imaging status at disease presentation i prognostic or if it can be used as a predictive biomarker for the depth or duration of response to erlotinib, and (b) if knowledge of the heterogeneity of 11C-erlotinib binding at the onset of EGFR TKI resistance can improve patient outcomes. The short term goals of the present study are designed to lay the foundation for a future clinical trial to test if knowing 11C-erlotinib status changes patient outcomes. The goals of the present pilot study are, (1) to determine the specific binding threshold that best differentiates EGFR mutant from EGFR wild-type tumors in newly diagnosed NSCLC; and (2) to determine the variability across patients in 11C- erlotinib specific binding at acquired resistance. If a threshold to separate patients can be identified at disease presentation, and variability across patients at resistance can be confirmed by this pilot study, then both measures will deserve to be evaluated in greater detail in a subsequent clinical trial for their respective values in improving patient outcomes. We will also pursue exploratory aims. These aims focus on (a) determining the sensitivity of the PET imaging analysis methods to error, (b) gaining preliminary data on the predictive value of erlotinib binding at disease presentation, and (c) evaluating the availability of 11C-erlotinib in patients with EGFR mutant NSCLC and brain metastases.